Traffic Flow Implications of Driverless Trucks - Microscopic Traffic Simulations using SUMO
(2020) In CODEN: LUTVDG/(TVTT-5319)/1-94/2020 VTVM01 20192Transport and Roads
- Abstract (Swedish)
- Godssektorn spås bli en tidig användare av självkörande fordonsteknik. Minskade drivmedelskostnader, ökad flexibilitet i ruttplaneringen och utökade drifttider är exempel på incitament för godsföretag att ställa om till en självkörande lastbilsflotta. Automatiseringen kommer mest sannolikt ske gradvis. I tidiga skeden spås självkörande lastbilar (AT:s) ha en konservativ körstil där körfältsbyten undviks i möjligaste mån och stora avstånd hålls till den omgivande trafiken. I ett senare skede i utvecklingen kommer körbeteendet bli mindre konservativt och vissa lastbilar kommer vara utrustade med V2V-kommunikationssystem.
I denna rapport används mikroskopiska trafiksimuleringar för att undersöka hur både självkörande lastbilar med en... (More) - Godssektorn spås bli en tidig användare av självkörande fordonsteknik. Minskade drivmedelskostnader, ökad flexibilitet i ruttplaneringen och utökade drifttider är exempel på incitament för godsföretag att ställa om till en självkörande lastbilsflotta. Automatiseringen kommer mest sannolikt ske gradvis. I tidiga skeden spås självkörande lastbilar (AT:s) ha en konservativ körstil där körfältsbyten undviks i möjligaste mån och stora avstånd hålls till den omgivande trafiken. I ett senare skede i utvecklingen kommer körbeteendet bli mindre konservativt och vissa lastbilar kommer vara utrustade med V2V-kommunikationssystem.
I denna rapport används mikroskopiska trafiksimuleringar för att undersöka hur både självkörande lastbilar med en konservativ körstil och med mer avancerade köregenskaper påverkar restider, vägkapacitet och utsläpp av koldioxid. När de självkörande lastbilarna simulerades i ett heterogent trafikflöde, visade det sig att den konservativa körstilen resulterade i ökade restider genom nätverket, samtidigt som köbildningar vid en påfartsramp påverkade den totala vägkapaciteten negativt. Självkörande lastbilar med mer utvecklade köregenskaper hade istället en positiv påverkan på både restider, vägkapacitet och utsläpp av koldioxid. Det är därför troligt självkörande lastbilar i ett tidigt skede i utvecklingen kan komma att påverka trafiksystemet negativt i flera anseenden. Längre fram i utvecklingen tros denna påverkan istället vara positiv.
Införandet av självkörande godstrafik talar för att en större andel gods kan köra nattetid då ingen personal behöver vara närvarande under transporten. Dessa lastbilar skulle också kunna köra långsammare än den omgivande trafiken för att spara energi. När denna typ av godstransporter simulerades, visade det sig att påverkan på restider var begränsad medan utsläppen av koldioxid ökade markant. Utsläppen ökade dessutom snabbt då trafikflödet på vägen ökade. För att säkerställa en begränsad miljöpåverkan är det därför viktigt att denna typ av trafik, då den introduceras, begränsas både tidsmässigt och geografiskt. (Less) - Abstract
- The truck sector is thought to be one of the first adopters of autonomous driving technology. Reduced fuel- and labour costs, increased flexibility in scheduling and increased hours of service, are all incentives for the freight companies to go autonomous. The automation of the truck fleet is likely to be gradual. At first, autonomous trucks (AT:s) are expected to have a conservative driving style, avoid lane-changes and keep large gaps to the surrounding traffic. As technology advances, the AT:s will eventually be capable of performing any type of driving maneuvers and to a certain extent make use of vehicle-to-vehicle communication systems (V2V).
In this thesis, the microscopic traffic simulation software SUMO is used to investigate... (More) - The truck sector is thought to be one of the first adopters of autonomous driving technology. Reduced fuel- and labour costs, increased flexibility in scheduling and increased hours of service, are all incentives for the freight companies to go autonomous. The automation of the truck fleet is likely to be gradual. At first, autonomous trucks (AT:s) are expected to have a conservative driving style, avoid lane-changes and keep large gaps to the surrounding traffic. As technology advances, the AT:s will eventually be capable of performing any type of driving maneuvers and to a certain extent make use of vehicle-to-vehicle communication systems (V2V).
In this thesis, the microscopic traffic simulation software SUMO is used to investigate how AT:s with a conservative driving style, and AT:s with enhanced driving capabilities, affect travel times, maximum road capacity and emissions of carbon dioxide. When simulating AT:s in a heterogeneous traffic flow, the conservative driving style turned out to have a negative effect on both maximum road capacity and average travel times. Congestion around an on-ramp that worked as a bottleneck, significantly increased when conservative AT:s were introduced. Total emissions of carbon dioxide, however, decreased. The AT with enhanced driving capabilities, showed a much better performance, with insignificant effects on travel times, increases in road capacity and decreases in emissions of carbon dioxide. It is therefore likely that, in early stages of AT deployment, the overall effect on traffic could be negative. Once the share of AT:s with enhanced driving capabilities increase, the effect is instead assumed to be positive.
The introduction of AT:s advocates an increased share of trucks at night, when there is free capacity available. In order to save energy, these AT:s would drive at much lower speeds than the surrounding traffic. Simulations show that nighttime AT traffic has limited effects on travel times but results in large increases in total emissions of carbon dioxide. The simulations showed that a very low traffic flow is crucial in order to ensure low emission levels. Introduction of nighttime AT traffic should therefore come with sufficient time- and route restrictions, in order to ensure ecological sustainability. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9014879
- author
- Erlandson, William LU
- supervisor
- organization
- course
- VTVM01 20192
- year
- 2020
- type
- H3 - Professional qualifications (4 Years - )
- subject
- keywords
- Traffic Simulation, Driverless, Autonomous, AT, SUMO, Freight
- publication/series
- CODEN: LUTVDG/(TVTT-5319)/1-94/2020
- report number
- 352
- ISSN
- 1653-1922
- language
- English
- id
- 9014879
- date added to LUP
- 2020-06-10 11:46:08
- date last changed
- 2020-06-10 11:46:08
@misc{9014879,
abstract = {{The truck sector is thought to be one of the first adopters of autonomous driving technology. Reduced fuel- and labour costs, increased flexibility in scheduling and increased hours of service, are all incentives for the freight companies to go autonomous. The automation of the truck fleet is likely to be gradual. At first, autonomous trucks (AT:s) are expected to have a conservative driving style, avoid lane-changes and keep large gaps to the surrounding traffic. As technology advances, the AT:s will eventually be capable of performing any type of driving maneuvers and to a certain extent make use of vehicle-to-vehicle communication systems (V2V).
In this thesis, the microscopic traffic simulation software SUMO is used to investigate how AT:s with a conservative driving style, and AT:s with enhanced driving capabilities, affect travel times, maximum road capacity and emissions of carbon dioxide. When simulating AT:s in a heterogeneous traffic flow, the conservative driving style turned out to have a negative effect on both maximum road capacity and average travel times. Congestion around an on-ramp that worked as a bottleneck, significantly increased when conservative AT:s were introduced. Total emissions of carbon dioxide, however, decreased. The AT with enhanced driving capabilities, showed a much better performance, with insignificant effects on travel times, increases in road capacity and decreases in emissions of carbon dioxide. It is therefore likely that, in early stages of AT deployment, the overall effect on traffic could be negative. Once the share of AT:s with enhanced driving capabilities increase, the effect is instead assumed to be positive.
The introduction of AT:s advocates an increased share of trucks at night, when there is free capacity available. In order to save energy, these AT:s would drive at much lower speeds than the surrounding traffic. Simulations show that nighttime AT traffic has limited effects on travel times but results in large increases in total emissions of carbon dioxide. The simulations showed that a very low traffic flow is crucial in order to ensure low emission levels. Introduction of nighttime AT traffic should therefore come with sufficient time- and route restrictions, in order to ensure ecological sustainability.}},
author = {{Erlandson, William}},
issn = {{1653-1922}},
language = {{eng}},
note = {{Student Paper}},
series = {{CODEN: LUTVDG/(TVTT-5319)/1-94/2020}},
title = {{Traffic Flow Implications of Driverless Trucks - Microscopic Traffic Simulations using SUMO}},
year = {{2020}},
}